The scheduling of molt in migratory birds

Abstract

We model the yearly cycle of small migratory birds to explain the variation in scheduling of complete molt, in particular why some birds molt immediately after breeding on the breeding grounds (summer molt) whereas others migrate to their wintering grounds before molt is initiated (winter molt). We employ the method of dynamic programming, because of its suitability for modelling life history traits. Feather quality and latitude entered the model as state variables and were assumed to affect survival rate and reproductive success. Migration and molt were assumed to be associated with increased mortality risks. By changing the parameters in the model we were able to generate most existing molt patterns, including summer and winter molt, biannual (summer and winter) molt, and molt migration. Our model suggests that the scheduling of molt is basically a result of a trade-off between having a high feather quality during breeding versus during the non-breeding period. A high impact of feather quality on survival rate in combination with low costs of molt resulted in biannual molt. Winter molt became more likely as the survival rateper se increased. A low seasonal amplitude in survival rate is a prerequisite for the occurrence of molt migration. Molt duration, migration costs and reproductive successper se were found to have no impact on the timing of molt. We also investigated the effect of benefits from prior occupancy at breeding and winter grounds.